基于软传感器技术的海水富营养化监测
|
摘要
藻类的繁殖生长状态很难用一种传感器在线、实时地直接测量出来。影响藻类生长的环境理化因子众多,这些环境因子之间又相互作用,海洋生态环境是一个高度复杂的非线性系统,很难用传统的机理建模方法来描述。文中用T-S模糊神经网络描述这种复杂的非线性系统,通过构造软传感器来测量藻类的繁殖生长状况。将叶绿素a的含量作为描述藻类生长状态的直接指标,并作为系统的输出变量,通过相关性分析,将影响藻类繁殖生长的主要环境因子作为系统的输入变量,通过对样本的学习训练,构造基于T-S模糊神经网络的软传感器模型。实验结果表明,这种软传感器模型能较好地描述可测环境因子与海水叶绿素a含量之间的非线性映射关系,验证了这种软传感器在监测海水水质异常变化时的有效性。
The state of algae growth is difficult to be measured by some kind of sensor directly.There are many environmental factors to affect the growth of algae,these factors have more internal functions affected each other,so the system is a complex nonlinear system,and it is difficult to be represented by mechanism model.In this paper,T-S fuzzy neural network is adopted to describe this kind of complex system.The state of algae growth can be measured by constructing the soft sensor.Content of chlorophyll-a can indicate the state of algae reproduction as the direct index,and it can be regarded as the output of the model system,some key environmental factors can be chosen as the input variables of the model system after correlation analysis.Soft sensor model based on T-S fuzzy neural network can be constructed by sample training.Experiment result illustrates that this kind of soft sensor model can well describe this nonlinear mapping relationship between the measurable environmental factors and the content of chlorophyll-a,and it can be effectively proved in early warning for the quality of seawater changing remarkably by this soft sensor.
The state of algae growth is difficult to be measured by some kind of sensor directly.There are many environmental factors to affect the growth of algae,these factors have more internal functions affected each other,so the system is a complex nonlinear system,and it is difficult to be represented by mechanism model.In this paper,T-S fuzzy neural network is adopted to describe this kind of complex system.The state of algae growth can be measured by constructing the soft sensor.Content of chlorophyll-a can indicate the state of algae reproduction as the direct index,and it can be regarded as the output of the model system,some key environmental factors can be chosen as the input variables of the model system after correlation analysis.Soft sensor model based on T-S fuzzy neural network can be constructed by sample training.Experiment result illustrates that this kind of soft sensor model can well describe this nonlinear mapping relationship between the measurable environmental factors and the content of chlorophyll-a,and it can be effectively proved in early warning for the quality of seawater changing remarkably by this soft sensor.
引文
[1]农吉夫,黄文宁.基于主成分分析的BP神经网络长期预报模型[J].广西师范学院学报:自然科学版,2008,25(4):46-51.NONG Ji-fu,HUANG Wen-ning.Long-tem prediction model of BP neural networks based on principle component analysis[J].Journal of Guangxi Teachers Education University:Natural Science Edition,2008,25(4):46-51.(in Chinese)
[2]杨建强,罗先香,丁德文,等.赤潮预测的人工神经网络方法初步研究[J].海洋科学进展,2003,25(4):318-323.
[3]Joseph H,HUANG Y,Mike D,et al.Neural network modelling of coastal algal blooms[J].Ecological Modelling,2003,159:179-201.
[4]Assefa M,Jayachandran K,ZHANG K Q.Modeling coastal eutrophication at florida bay using neural etworks[J].Journal ofCoastal Research,2008,24(28):190-196.
[5]王洪礼,葛根,李悦雷.基于模糊神经网络(FNN)的赤潮预警预测研究[J].海洋通报,2006,25(4):36-41.WANG Hong-li,GE Gen,LI Yue-lei.Research on the prediction of red tide based on the fuzzy neural network[J].MarineScience Bulletin,2006,25(4):36-41.(in Chinese)
[6]刘载文,吕思颖,王小艺,等.河湖水华预测方法研究[J].水资源保,2008,24(5):42-47.LIU Zai-wen,LV Si-ying,WANG Xiao-yi.et al.Forecast methods for algal bloom in rivers and lakes[J].Water ResourcesProtection,2008,24(5):42-47.(in Chinese)
[7]李永振.利用主成分分析法研究东北地震活动异常[J].地震研究,2010,3(2):164-168.LI Yong-zhen.Using the primary component analysis earthquake parameters reduction[J].Journal of Seismological Research,2010,3(2):164-168.(in Chinese)
[8]张颖.改进的T-S模糊神经网络在化工软测量中的应用[J].电子测量与仪器学报,2010,24(6):585-589.ZHANG Ying.Improved T-S fuzzy neural network applied in soft sensing of chemical industry[J].Journal of Electronic Measure-mentand Instrument,2010,24(6):585-589.(in Chinese)
[9]高强,宋祎,杜忠晓.基于T-S模糊神经网络的信息融合在赤潮预测预警中的应用[J].海洋技术,2006,25(2):103-105.GAO Qiang,SONG Yi,DU Zhong-xiao.Information fusion technology based on T-S fuzzy neiual networks for red tide prediction[J].Ocean Technology,2006,25(2):103-105.(in Chinese)
[2]杨建强,罗先香,丁德文,等.赤潮预测的人工神经网络方法初步研究[J].海洋科学进展,2003,25(4):318-323.
[3]Joseph H,HUANG Y,Mike D,et al.Neural network modelling of coastal algal blooms[J].Ecological Modelling,2003,159:179-201.
[4]Assefa M,Jayachandran K,ZHANG K Q.Modeling coastal eutrophication at florida bay using neural etworks[J].Journal ofCoastal Research,2008,24(28):190-196.
[5]王洪礼,葛根,李悦雷.基于模糊神经网络(FNN)的赤潮预警预测研究[J].海洋通报,2006,25(4):36-41.WANG Hong-li,GE Gen,LI Yue-lei.Research on the prediction of red tide based on the fuzzy neural network[J].MarineScience Bulletin,2006,25(4):36-41.(in Chinese)
[6]刘载文,吕思颖,王小艺,等.河湖水华预测方法研究[J].水资源保,2008,24(5):42-47.LIU Zai-wen,LV Si-ying,WANG Xiao-yi.et al.Forecast methods for algal bloom in rivers and lakes[J].Water ResourcesProtection,2008,24(5):42-47.(in Chinese)
[7]李永振.利用主成分分析法研究东北地震活动异常[J].地震研究,2010,3(2):164-168.LI Yong-zhen.Using the primary component analysis earthquake parameters reduction[J].Journal of Seismological Research,2010,3(2):164-168.(in Chinese)
[8]张颖.改进的T-S模糊神经网络在化工软测量中的应用[J].电子测量与仪器学报,2010,24(6):585-589.ZHANG Ying.Improved T-S fuzzy neural network applied in soft sensing of chemical industry[J].Journal of Electronic Measure-mentand Instrument,2010,24(6):585-589.(in Chinese)
[9]高强,宋祎,杜忠晓.基于T-S模糊神经网络的信息融合在赤潮预测预警中的应用[J].海洋技术,2006,25(2):103-105.GAO Qiang,SONG Yi,DU Zhong-xiao.Information fusion technology based on T-S fuzzy neiual networks for red tide prediction[J].Ocean Technology,2006,25(2):103-105.(in Chinese)
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心